Preferably, an error correction code used for error correction is evenly inserted to bits in a case of general data different from image data, because a significance of each bit is the same. However, a significance of bits composing the image data may differ. In a case of the image data, pixels expressing the image may be composed of data of various bits. For example, in a case of pixels expressed with RGB (Red, Green, and Blue), each color may be expressed with total 24-bits being comprised of three 8-bits. 8-bits composing each color may express 256 colors. 8-bit color pixel may have a similar color as that of an adjacent 8-bit color pixel. When errors occur in lower bits according to the standardized RGB color coordinates, a difference in the colors may be insignificantly small, however when errors occur in higher bits, the difference in the colors may be significantly large.
In a case of bits expressing colors as described above, the significance may differ for each bit. Disadvantageously, according to a conventional image transmission method, the same significance is applied to each of the bits without considering the significance for each bit, and the error correction code is inserted to the bits having the same significance.